----------------------------------------------------------------------------------
      name:  <unnamed>
       log:  C:\research\china\decentralization\restat_data\results\table4.log
  log type:  text
 opened on:  22 Jul 2016, 09:37:46

. 
. 
. /*
> *instructions for outreg;
> local instruct_ols "ctitle(OLS) dec(4)  auto(2) symbol(***,**,*) se"; 
> local instruct_iv "ctitle(IV) dec(4)  auto(2) symbol(***,**,*) se addstat(First 
> stage F, e(widstat))"; 
> local instruct_iv_overid "ctitle(IV) dec(4)  auto(2) symbol(***,**,*)  se addsta
> t(Overid,e(jp),First stage F, e(widstat))"; 
> */
. 
. 
. ********************************************************************; 
. ***     TABLE 4 pop growth 1990_2010 -- IV          
. ********************************************************************; 
. 
. *stripped down specification
. ivreg2 D_censuspop9010_cp (all_road_2010_rays=road_1962_rays) if D0==1,  first f
> first robust cluster(province05)

First-stage regressions
-----------------------


First-stage regression of all_road_2010_rays:

Statistics robust to heteroskedasticity and clustering on province05
Number of obs =                    257
Number of clusters (province05) =     26
--------------------------------------------------------------------------------
               |               Robust
all_road_201~s |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------+----------------------------------------------------------------
road_1962_rays |   .5232738   .0869323     6.02   0.000     .3520771    .6944704
         _cons |   2.744466    .234125    11.72   0.000     2.283401    3.205531
--------------------------------------------------------------------------------
F test of excluded instruments:
  F(  1,    25) =    36.23
  Prob > F      =   0.0000
Sanderson-Windmeijer multivariate F test of excluded instruments:
  F(  1,    25) =    36.23
  Prob > F      =   0.0000



Summary results for first-stage regressions
-------------------------------------------

                                           (Underid)            (Weak id)
Variable     | F(  1,    25)  P-val | SW Chi-sq(  1) P-val | SW F(  1,    25)
all_road_201 |      36.23    0.0000 |       37.83   0.0000 |       36.23

NB: first-stage test statistics cluster-robust

Stock-Yogo weak ID F test critical values for single endogenous regressor:
                                   10% maximal IV size             16.38
                                   15% maximal IV size              8.96
                                   20% maximal IV size              6.66
                                   25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for i.i.d. errors only.

Underidentification test
Ho: matrix of reduced form coefficients has rank=K1-1 (underidentified)
Ha: matrix has rank=K1 (identified)
Kleibergen-Paap rk LM statistic          Chi-sq(1)=11.17    P-val=0.0008

Weak identification test
Ho: equation is weakly identified
Cragg-Donald Wald F statistic                                      37.09
Kleibergen-Paap Wald rk F statistic                                36.23

Stock-Yogo weak ID test critical values for K1=1 and L1=1:
                                   10% maximal IV size             16.38
                                   15% maximal IV size              8.96
                                   20% maximal IV size              6.66
                                   25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.

Weak-instrument-robust inference
Tests of joint significance of endogenous regressors B1 in main equation
Ho: B1=0 and orthogonality conditions are valid
Anderson-Rubin Wald test           F(1,25)=        0.13     P-val=0.7254
Anderson-Rubin Wald test           Chi-sq(1)=      0.13     P-val=0.7166
Stock-Wright LM S statistic        Chi-sq(1)=      0.26     P-val=0.6084

NB: Underidentification, weak identification and weak-identification-robust
    test statistics cluster-robust

Number of clusters             N_clust  =         26
Number of observations               N  =        257
Number of regressors                 K  =          2
Number of endogenous regressors      K1 =          1
Number of instruments                L  =          2
Number of excluded instruments       L1 =          1

IV (2SLS) estimation
--------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on province05

Number of clusters (province05) =     26              Number of obs =      257
                                                      F(  1,    25) =     0.13
                                                      Prob > F      =   0.7261
Total (centered) SS     =  24.19405425                Centered R2   =  -0.0077
Total (uncentered) SS   =   67.2418726                Uncentered R2 =   0.6374
Residual SS             =  24.37982258                Root MSE      =     .308

---------------------------------------------------------------------------------
                |               Robust
D_censu~9010_cp |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
all_road_2010~s |  -.0067411   .0186204    -0.36   0.717    -.0432364    .0297543
          _cons |   .4349481   .0971293     4.48   0.000     .2445781    .6253181
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             11.169
                                                   Chi-sq(1) P-val =    0.0008
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):               37.095
                         (Kleibergen-Paap rk Wald F statistic):         36.232
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         all_road_2010_rays
Excluded instruments: road_1962_rays
------------------------------------------------------------------------------

. outreg2 using "table4", excel  dec(4)  auto(2) symbol(***,**,*)  se addstat(Firs
> t stage F, e(widstat))  replace
table4.xml
dir : seeout

. *base specification
. ivreg2 D_censuspop9010_cp larea_cc90 larea_pf05 province_capitalPlus lcensuspop1
> 982_pref frac_highedu1982_pref share_emp_man1982_pref (all_road_2010_rays D_cens
> uspop9010_pref=road_1962_rays ruralMigp00) if D0==1,  first ffirst robust cluste
> r(province05)

Unable to store first-stage regression of D_censuspop9010_pref.


First-stage regressions
-----------------------

Unable to display all first-stage regressions.
There may be insufficient room to store results using -estimates store-,
or names of endogenous regressors may be too long to store the results.
Try dropping one or more estimation results using -estimates drop-,
using the -savefprefix- option, or using shorter variable names.


First-stage regression of all_road_2010_rays:

Statistics robust to heteroskedasticity and clustering on province05
Number of obs =                    257
Number of clusters (province05) =     26
---------------------------------------------------------------------------------
                |               Robust
all_road_2010~s |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
 road_1962_rays |   .3725221   .0821979     4.53   0.000     .2106271    .5344171
    ruralMigp00 |   2.16e-06   1.10e-06     1.97   0.050    -2.42e-09    4.33e-06
     larea_cc90 |   .1261369    .121594     1.04   0.301    -.1133517    .3656255
     larea_pf05 |   .0435905   .1989704     0.22   0.827    -.3482968    .4354778
province_capi~s |   1.466491   .5226598     2.81   0.005     .4370726    2.495909
lcensuspop198~f |   .7673455   .1800237     4.26   0.000     .4127752    1.121916
frac_hig~2_pref |   5.695243   2.814337     2.02   0.044     .1521928    11.23829
sha~an1982_pref |  -4.039685   2.086653    -1.94   0.054    -8.149505    .0701356
          _cons |  -10.13675   2.707129    -3.74   0.000    -15.46865   -4.804855
---------------------------------------------------------------------------------
F test of excluded instruments:
  F(  2,    25) =    13.62
  Prob > F      =   0.0001
Sanderson-Windmeijer multivariate F test of excluded instruments:
  F(  1,    25) =    27.49
  Prob > F      =   0.0000



Summary results for first-stage regressions
-------------------------------------------

                                           (Underid)            (Weak id)
Variable     | F(  2,    25)  P-val | SW Chi-sq(  1) P-val | SW F(  1,    25)
all_road_201 |      13.62    0.0001 |       29.52   0.0000 |       27.49
D_censuspop9 |       4.13    0.0281 |       35.57   0.0000 |       33.13

NB: first-stage test statistics cluster-robust

Stock-Yogo weak ID F test critical values for single endogenous regressor:
                                   10% maximal IV size             19.93
                                   15% maximal IV size             11.59
                                   20% maximal IV size              8.75
                                   25% maximal IV size              7.25
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for i.i.d. errors only.

Underidentification test
Ho: matrix of reduced form coefficients has rank=K1-1 (underidentified)
Ha: matrix has rank=K1 (identified)
Kleibergen-Paap rk LM statistic          Chi-sq(1)=11.56    P-val=0.0007

Weak identification test
Ho: equation is weakly identified
Cragg-Donald Wald F statistic                                      10.50
Kleibergen-Paap Wald rk F statistic                                13.09

Stock-Yogo weak ID test critical values for K1=2 and L1=2:
                                   10% maximal IV size              7.03
                                   15% maximal IV size              4.58
                                   20% maximal IV size              3.95
                                   25% maximal IV size              3.63
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.

Weak-instrument-robust inference
Tests of joint significance of endogenous regressors B1 in main equation
Ho: B1=0 and orthogonality conditions are valid
Anderson-Rubin Wald test           F(2,25)=        4.37     P-val=0.0237
Anderson-Rubin Wald test           Chi-sq(2)=      9.37     P-val=0.0092
Stock-Wright LM S statistic        Chi-sq(2)=      9.68     P-val=0.0079

NB: Underidentification, weak identification and weak-identification-robust
    test statistics cluster-robust

Number of clusters             N_clust  =         26
Number of observations               N  =        257
Number of regressors                 K  =          9
Number of endogenous regressors      K1 =          2
Number of instruments                L  =          9
Number of excluded instruments       L1 =          2

IV (2SLS) estimation
--------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on province05

Number of clusters (province05) =     26              Number of obs =      257
                                                      F(  8,    25) =    21.21
                                                      Prob > F      =   0.0000
Total (centered) SS     =  24.19405425                Centered R2   =   0.4525
Total (uncentered) SS   =   67.2418726                Uncentered R2 =   0.8030
Residual SS             =  13.24583226                Root MSE      =     .227

---------------------------------------------------------------------------------
                |               Robust
D_censu~9010_cp |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
all_road_2010~s |  -.0422987   .0222676    -1.90   0.057    -.0859423    .0013449
D_cen~9010_pref |   .8123965   .1388716     5.85   0.000     .5402131     1.08458
     larea_cc90 |  -.1178236   .0191204    -6.16   0.000    -.1552988   -.0803484
     larea_pf05 |   .0507944   .0178213     2.85   0.004     .0158652    .0857235
province_capi~s |    .175135   .0723817     2.42   0.016     .0332694    .3170006
lcensuspop198~f |   .1100811   .0365383     3.01   0.003     .0384673    .1816949
frac_hig~2_pref |  -.3789841   .3414735    -1.11   0.267     -1.04826    .2902917
sha~an1982_pref |  -.2845332   .2543698    -1.12   0.263    -.7830889    .2140225
          _cons |  -.7534542   .5666954    -1.33   0.184    -1.864157    .3572485
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             11.560
                                                   Chi-sq(1) P-val =    0.0007
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):               10.496
                         (Kleibergen-Paap rk Wald F statistic):         13.093
Stock-Yogo weak ID test critical values: 10% maximal IV size              7.03
                                         15% maximal IV size              4.58
                                         20% maximal IV size              3.95
                                         25% maximal IV size              3.63
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         all_road_2010_rays D_censuspop9010_pref
Included instruments: larea_cc90 larea_pf05 province_capitalPlus
                      lcensuspop1982_pref frac_highedu1982_pref
                      share_emp_man1982_pref
Excluded instruments: road_1962_rays ruralMigp00
------------------------------------------------------------------------------

. outreg2 using "table4", excel  dec(4)  auto(2) symbol(***,**,*)  se addstat(Firs
> t stage F, e(widstat))  append
table4.xml
dir : seeout

. *add km roads outside CC
. ivreg2 D_censuspop9010_cp larea_cc90 larea_pf05 province_capitalPlus lcensuspop1
> 982_pref frac_highedu1982_pref share_emp_man1982_pref (all_road_2010_rays D_cens
> uspop9010_pref lall_road_2010_km_pfcp=road_1962_rays ruralMigp00 lroad_1962_km_p
> fcp) if D0==1, first ffirst robust cluster(province05)

Unable to store first-stage regression of D_censuspop9010_pref.


Unable to store first-stage regression of lall_road_2010_km_pfcp.


First-stage regressions
-----------------------

Unable to display all first-stage regressions.
There may be insufficient room to store results using -estimates store-,
or names of endogenous regressors may be too long to store the results.
Try dropping one or more estimation results using -estimates drop-,
using the -savefprefix- option, or using shorter variable names.


First-stage regression of all_road_2010_rays:

Statistics robust to heteroskedasticity and clustering on province05
Number of obs =                    257
Number of clusters (province05) =     26
---------------------------------------------------------------------------------
                |               Robust
all_road_2010~s |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
 road_1962_rays |    .377663   .0804242     4.70   0.000     .2192582    .5360677
    ruralMigp00 |   2.15e-06   1.10e-06     1.96   0.051    -1.23e-08    4.31e-06
lroad_1962_km~p |  -.0527682   .1363956    -0.39   0.699    -.3214149    .2158785
     larea_cc90 |   .1151668   .1317885     0.87   0.383    -.1444058    .3747394
     larea_pf05 |   .0977543   .2589832     0.38   0.706    -.4123429    .6078515
province_capi~s |   1.450589   .5039266     2.88   0.004     .4580475     2.44313
lcensuspop198~f |   .7857452   .2018348     3.89   0.000     .3882084    1.183282
frac_hig~2_pref |   5.734974   2.824903     2.03   0.043     .1710022    11.29894
sha~an1982_pref |  -4.010645   2.027862    -1.98   0.049    -8.004752   -.0165376
          _cons |  -10.57132   3.222833    -3.28   0.001    -16.91906   -4.223576
---------------------------------------------------------------------------------
F test of excluded instruments:
  F(  3,    25) =     9.37
  Prob > F      =   0.0002
Sanderson-Windmeijer multivariate F test of excluded instruments:
  F(  1,    25) =    31.10
  Prob > F      =   0.0000



Summary results for first-stage regressions
-------------------------------------------

                                           (Underid)            (Weak id)
Variable     | F(  3,    25)  P-val | SW Chi-sq(  1) P-val | SW F(  1,    25)
all_road_201 |       9.37    0.0002 |       33.53   0.0000 |       31.10
D_censuspop9 |       4.96    0.0077 |       35.96   0.0000 |       33.36
lall_road_20 |      12.65    0.0000 |       10.50   0.0012 |        9.74

NB: first-stage test statistics cluster-robust

Stock-Yogo weak ID F test critical values for single endogenous regressor:
                                    5% maximal IV relative bias    13.91
                                   10% maximal IV relative bias     9.08
                                   20% maximal IV relative bias     6.46
                                   30% maximal IV relative bias     5.39
                                   10% maximal IV size             22.30
                                   15% maximal IV size             12.83
                                   20% maximal IV size              9.54
                                   25% maximal IV size              7.80
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for i.i.d. errors only.

Underidentification test
Ho: matrix of reduced form coefficients has rank=K1-1 (underidentified)
Ha: matrix has rank=K1 (identified)
Kleibergen-Paap rk LM statistic          Chi-sq(1)=11.65    P-val=0.0006

Weak identification test
Ho: equation is weakly identified
Cragg-Donald Wald F statistic                                       6.99
Kleibergen-Paap Wald rk F statistic                                 8.81

Stock-Yogo weak ID test critical values for K1=3 and L1=3:
                                                               <not available>

Weak-instrument-robust inference
Tests of joint significance of endogenous regressors B1 in main equation
Ho: B1=0 and orthogonality conditions are valid
Anderson-Rubin Wald test           F(3,25)=        7.70     P-val=0.0008
Anderson-Rubin Wald test           Chi-sq(3)=     24.91     P-val=0.0000
Stock-Wright LM S statistic        Chi-sq(3)=     12.98     P-val=0.0047

NB: Underidentification, weak identification and weak-identification-robust
    test statistics cluster-robust

Number of clusters             N_clust  =         26
Number of observations               N  =        257
Number of regressors                 K  =         10
Number of endogenous regressors      K1 =          3
Number of instruments                L  =         10
Number of excluded instruments       L1 =          3

IV (2SLS) estimation
--------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on province05

Number of clusters (province05) =     26              Number of obs =      257
                                                      F(  9,    25) =    18.16
                                                      Prob > F      =   0.0000
Total (centered) SS     =  24.19405425                Centered R2   =   0.4434
Total (uncentered) SS   =   67.2418726                Uncentered R2 =   0.7997
Residual SS             =  13.46587244                Root MSE      =    .2289

---------------------------------------------------------------------------------
                |               Robust
D_censu~9010_cp |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
all_road_2010~s |  -.0448366   .0227789    -1.97   0.049    -.0894825   -.0001907
D_cen~9010_pref |   .7554976   .1800656     4.20   0.000     .4025756     1.10842
lall_road_201~p |    .088477   .0797184     1.11   0.267    -.0677683    .2447223
     larea_cc90 |  -.0966007   .0224895    -4.30   0.000    -.1406792   -.0525222
     larea_pf05 |  -.0334828   .0848276    -0.39   0.693    -.1997419    .1327763
province_capi~s |    .186447   .0733098     2.54   0.011     .0427625    .3301315
lcensuspop198~f |   .0784063   .0548587     1.43   0.153    -.0291147    .1859273
frac_hig~2_pref |   -.418469   .3489334    -1.20   0.230    -1.102366    .2654279
sha~an1982_pref |  -.2651717    .271722    -0.98   0.329    -.7977371    .2673936
          _cons |  -.1753885   .9524634    -0.18   0.854    -2.042182    1.691405
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             11.647
                                                   Chi-sq(1) P-val =    0.0006
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):                6.994
                         (Kleibergen-Paap rk Wald F statistic):          8.806
Stock-Yogo weak ID test critical values:                       <not available>
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         all_road_2010_rays D_censuspop9010_pref
                      lall_road_2010_km_pfcp
Included instruments: larea_cc90 larea_pf05 province_capitalPlus
                      lcensuspop1982_pref frac_highedu1982_pref
                      share_emp_man1982_pref
Excluded instruments: road_1962_rays ruralMigp00 lroad_1962_km_pfcp
------------------------------------------------------------------------------

. outreg2 using "table4", excel  dec(4)  auto(2) symbol(***,**,*)  se addstat(Firs
> t stage F, e(widstat)) append
table4.xml
dir : seeout

. *add radial rr instead
. ivreg2 D_censuspop9010_cp larea_cc90 larea_pf05 province_capitalPlus lcensuspop1
> 982_pref frac_highedu1982_pref share_emp_man1982_pref (all_road_2010_rays D_cens
> uspop9010_pref rail_2010_rays=road_1962_rays ruralMigp00 rail_1962_rays) if D0==
> 1, first ffirst robust cluster(province05)

Unable to store first-stage regression of D_censuspop9010_pref.


First-stage regressions
-----------------------

Unable to display all first-stage regressions.
There may be insufficient room to store results using -estimates store-,
or names of endogenous regressors may be too long to store the results.
Try dropping one or more estimation results using -estimates drop-,
using the -savefprefix- option, or using shorter variable names.


First-stage regression of all_road_2010_rays:

Statistics robust to heteroskedasticity and clustering on province05
Number of obs =                    257
Number of clusters (province05) =     26
---------------------------------------------------------------------------------
                |               Robust
all_road_2010~s |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
 road_1962_rays |    .361789   .0846058     4.28   0.000     .1951482    .5284299
    ruralMigp00 |   2.10e-06   1.10e-06     1.90   0.058    -7.26e-08    4.27e-06
 rail_1962_rays |   .1783403   .1078123     1.65   0.099    -.0340083    .3906889
     larea_cc90 |   .1426515   .1180601     1.21   0.228    -.0898814    .3751843
     larea_pf05 |   .0449543   .2000875     0.22   0.822     -.349141    .4390496
province_capi~s |   1.393904   .5102111     2.73   0.007     .3889848    2.398824
lcensuspop198~f |   .7034736   .1918043     3.67   0.000     .3256931    1.081254
frac_hig~2_pref |   4.112019   2.590508     1.59   0.114    -.9902831    9.214322
sha~an1982_pref |  -4.261199   2.095602    -2.03   0.043    -8.388728   -.1336697
          _cons |  -9.281219   2.852358    -3.25   0.001    -14.89927   -3.663171
---------------------------------------------------------------------------------
F test of excluded instruments:
  F(  3,    25) =     9.34
  Prob > F      =   0.0003
Sanderson-Windmeijer multivariate F test of excluded instruments:
  F(  1,    25) =    21.58
  Prob > F      =   0.0001


First-stage regression of rail_2010_rays:

Statistics robust to heteroskedasticity and clustering on province05
Number of obs =                    257
Number of clusters (province05) =     26
---------------------------------------------------------------------------------
                |               Robust
 rail_2010_rays |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
 road_1962_rays |   .0212813   .0345853     0.62   0.539    -.0468385    .0894011
    ruralMigp00 |  -9.97e-07   7.08e-07    -1.41   0.161    -2.39e-06    3.98e-07
 rail_1962_rays |   .3738095   .0533411     7.01   0.000      .268748     .478871
     larea_cc90 |  -.0070261    .091546    -0.08   0.939    -.1873364    .1732842
     larea_pf05 |  -.0539781   .1656476    -0.33   0.745      -.38024    .2722839
province_capi~s |    .206971   .2336492     0.89   0.377     -.253228    .6671699
lcensuspop198~f |   .4354343   .1324146     3.29   0.001     .1746286      .69624
frac_hig~2_pref |   4.431509   2.123627     2.09   0.038     .2487816    8.614236
sha~an1982_pref |  -1.344496   .6320362    -2.13   0.034    -2.589364   -.0996282
          _cons |  -4.846523   2.064375    -2.35   0.020    -8.912547   -.7804988
---------------------------------------------------------------------------------
F test of excluded instruments:
  F(  3,    25) =    25.13
  Prob > F      =   0.0000
Sanderson-Windmeijer multivariate F test of excluded instruments:
  F(  1,    25) =    38.95
  Prob > F      =   0.0000



Summary results for first-stage regressions
-------------------------------------------

                                           (Underid)            (Weak id)
Variable     | F(  3,    25)  P-val | SW Chi-sq(  1) P-val | SW F(  1,    25)
all_road_201 |       9.34    0.0003 |       23.26   0.0000 |       21.58
D_censuspop9 |       4.11    0.0169 |       31.95   0.0000 |       29.64
rail_2010_ra |      25.13    0.0000 |       41.99   0.0000 |       38.95

NB: first-stage test statistics cluster-robust

Stock-Yogo weak ID F test critical values for single endogenous regressor:
                                    5% maximal IV relative bias    13.91
                                   10% maximal IV relative bias     9.08
                                   20% maximal IV relative bias     6.46
                                   30% maximal IV relative bias     5.39
                                   10% maximal IV size             22.30
                                   15% maximal IV size             12.83
                                   20% maximal IV size              9.54
                                   25% maximal IV size              7.80
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for i.i.d. errors only.

Underidentification test
Ho: matrix of reduced form coefficients has rank=K1-1 (underidentified)
Ha: matrix has rank=K1 (identified)
Kleibergen-Paap rk LM statistic          Chi-sq(1)=9.27     P-val=0.0023

Weak identification test
Ho: equation is weakly identified
Cragg-Donald Wald F statistic                                       5.32
Kleibergen-Paap Wald rk F statistic                                 7.02

Stock-Yogo weak ID test critical values for K1=3 and L1=3:
                                                               <not available>

Weak-instrument-robust inference
Tests of joint significance of endogenous regressors B1 in main equation
Ho: B1=0 and orthogonality conditions are valid
Anderson-Rubin Wald test           F(3,25)=        3.14     P-val=0.0431
Anderson-Rubin Wald test           Chi-sq(3)=     10.15     P-val=0.0173
Stock-Wright LM S statistic        Chi-sq(3)=      9.84     P-val=0.0199

NB: Underidentification, weak identification and weak-identification-robust
    test statistics cluster-robust

Number of clusters             N_clust  =         26
Number of observations               N  =        257
Number of regressors                 K  =         10
Number of endogenous regressors      K1 =          3
Number of instruments                L  =         10
Number of excluded instruments       L1 =          3

IV (2SLS) estimation
--------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on province05

Number of clusters (province05) =     26              Number of obs =      257
                                                      F(  9,    25) =    20.23
                                                      Prob > F      =   0.0000
Total (centered) SS     =  24.19405425                Centered R2   =   0.4469
Total (uncentered) SS   =   67.2418726                Uncentered R2 =   0.8010
Residual SS             =  13.38241858                Root MSE      =    .2282

---------------------------------------------------------------------------------
                |               Robust
D_censu~9010_cp |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
all_road_2010~s |  -.0412392   .0246014    -1.68   0.094     -.089457    .0069787
D_cen~9010_pref |   .7974965   .1656862     4.81   0.000     .4727575    1.122236
 rail_2010_rays |  -.0104888   .0485386    -0.22   0.829    -.1056227    .0846451
     larea_cc90 |  -.1187856   .0205051    -5.79   0.000    -.1589749   -.0785963
     larea_pf05 |   .0495115   .0194369     2.55   0.011     .0114159    .0876071
province_capi~s |   .1798078   .0765698     2.35   0.019     .0297338    .3298818
lcensuspop198~f |   .1140457    .030362     3.76   0.000     .0545372    .1735541
frac_hig~2_pref |  -.3061648   .5069959    -0.60   0.546    -1.299859     .687529
sha~an1982_pref |  -.2880694   .2464661    -1.17   0.242    -.7711341    .1949952
          _cons |  -.7846209    .515102    -1.52   0.128    -1.794202    .2249606
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):              9.268
                                                   Chi-sq(1) P-val =    0.0023
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):                5.322
                         (Kleibergen-Paap rk Wald F statistic):          7.018
Stock-Yogo weak ID test critical values:                       <not available>
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         all_road_2010_rays D_censuspop9010_pref rail_2010_rays
Included instruments: larea_cc90 larea_pf05 province_capitalPlus
                      lcensuspop1982_pref frac_highedu1982_pref
                      share_emp_man1982_pref
Excluded instruments: road_1962_rays ruralMigp00 rail_1962_rays
------------------------------------------------------------------------------

. outreg2 using "table4", excel  dec(4)  auto(2) symbol(***,**,*)  se addstat(Firs
> t stage F, e(widstat)) append
table4.xml
dir : seeout

. *ring-road only
. ivreg2 D_censuspop9010_cp larea_cc90 larea_pf05 province_capitalPlus lcensuspop1
> 982_pref frac_highedu1982_pref share_emp_man1982_pref (po_s_all_road_2010_ringxd
> a D_censuspop9010_pref = po_s_road_1962_ringxda ruralMigp00 ) if D0==1, first ff
> irst robust cluster(province05)

Unable to store first-stage regression of po_s_all_road_2010_ringxda.


Unable to store first-stage regression of D_censuspop9010_pref.


First-stage regressions
-----------------------

Unable to display all first-stage regressions.
There may be insufficient room to store results using -estimates store-,
or names of endogenous regressors may be too long to store the results.
Try dropping one or more estimation results using -estimates drop-,
using the -savefprefix- option, or using shorter variable names.



Summary results for first-stage regressions
-------------------------------------------

                                           (Underid)            (Weak id)
Variable     | F(  2,    25)  P-val | SW Chi-sq(  1) P-val | SW F(  1,    25)
po_s_all_roa |       8.23    0.0018 |       14.89   0.0001 |       13.87
D_censuspop9 |       3.33    0.0521 |       25.51   0.0000 |       23.76

NB: first-stage test statistics cluster-robust

Stock-Yogo weak ID F test critical values for single endogenous regressor:
                                   10% maximal IV size             19.93
                                   15% maximal IV size             11.59
                                   20% maximal IV size              8.75
                                   25% maximal IV size              7.25
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for i.i.d. errors only.

Underidentification test
Ho: matrix of reduced form coefficients has rank=K1-1 (underidentified)
Ha: matrix has rank=K1 (identified)
Kleibergen-Paap rk LM statistic          Chi-sq(1)=5.17     P-val=0.0229

Weak identification test
Ho: equation is weakly identified
Cragg-Donald Wald F statistic                                       9.45
Kleibergen-Paap Wald rk F statistic                                 7.13

Stock-Yogo weak ID test critical values for K1=2 and L1=2:
                                   10% maximal IV size              7.03
                                   15% maximal IV size              4.58
                                   20% maximal IV size              3.95
                                   25% maximal IV size              3.63
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.

Weak-instrument-robust inference
Tests of joint significance of endogenous regressors B1 in main equation
Ho: B1=0 and orthogonality conditions are valid
Anderson-Rubin Wald test           F(2,25)=        3.82     P-val=0.0356
Anderson-Rubin Wald test           Chi-sq(2)=      8.20     P-val=0.0165
Stock-Wright LM S statistic        Chi-sq(2)=      8.41     P-val=0.0149

NB: Underidentification, weak identification and weak-identification-robust
    test statistics cluster-robust

Number of clusters             N_clust  =         26
Number of observations               N  =        257
Number of regressors                 K  =          9
Number of endogenous regressors      K1 =          2
Number of instruments                L  =          9
Number of excluded instruments       L1 =          2

IV (2SLS) estimation
--------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on province05

Number of clusters (province05) =     26              Number of obs =      257
                                                      F(  8,    25) =    21.06
                                                      Prob > F      =   0.0000
Total (centered) SS     =  24.19405425                Centered R2   =   0.3807
Total (uncentered) SS   =   67.2418726                Uncentered R2 =   0.7772
Residual SS             =  14.98324786                Root MSE      =    .2415

---------------------------------------------------------------------------------
                |               Robust
D_censu~9010_cp |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
po_s_all_road.. |  -.1872636   .0915789    -2.04   0.041    -.3667549   -.0077724
D_cen~9010_pref |   .6227582   .1478662     4.21   0.000     .3329457    .9125706
     larea_cc90 |  -.1619957   .0294504    -5.50   0.000    -.2197174    -.104274
     larea_pf05 |   .0388412   .0282975     1.37   0.170    -.0166209    .0943033
province_capi~s |   .1393413   .0573547     2.43   0.015      .026928    .2517545
lcensuspop198~f |    .069869   .0349123     2.00   0.045     .0014421    .1382958
frac_hig~2_pref |  -.4778669   .4515642    -1.06   0.290    -1.362916    .4071826
sha~an1982_pref |  -.0415398   .2485957    -0.17   0.867    -.5287785    .4456988
          _cons |   .1748527   .5206988     0.34   0.737    -.8456982    1.195404
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):              5.175
                                                   Chi-sq(1) P-val =    0.0229
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):                9.454
                         (Kleibergen-Paap rk Wald F statistic):          7.125
Stock-Yogo weak ID test critical values: 10% maximal IV size              7.03
                                         15% maximal IV size              4.58
                                         20% maximal IV size              3.95
                                         25% maximal IV size              3.63
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         po_s_all_road_2010_ringxda D_censuspop9010_pref
Included instruments: larea_cc90 larea_pf05 province_capitalPlus
                      lcensuspop1982_pref frac_highedu1982_pref
                      share_emp_man1982_pref
Excluded instruments: po_s_road_1962_ringxda ruralMigp00
------------------------------------------------------------------------------

. outreg2 using "table4", excel  dec(4)  auto(2) symbol(***,**,*)  se addstat(Firs
> t stage F, e(widstat)) append
table4.xml
dir : seeout

. *radial + ring-roads
. ivreg2 D_censuspop9010_cp larea_cc90 larea_pf05 province_capitalPlus lcensuspop1
> 982_pref frac_highedu1982_pref share_emp_man1982_pref (all_road_2010_rays po_s_a
> ll_road_2010_ringxda D_censuspop9010_pref = road_1962_rays po_s_road_1962_ringxd
> a ruralMigp00 ) if D0==1, first ffirst robust cluster(province05)

Unable to store first-stage regression of po_s_all_road_2010_ringxda.


Unable to store first-stage regression of D_censuspop9010_pref.


First-stage regressions
-----------------------

Unable to display all first-stage regressions.
There may be insufficient room to store results using -estimates store-,
or names of endogenous regressors may be too long to store the results.
Try dropping one or more estimation results using -estimates drop-,
using the -savefprefix- option, or using shorter variable names.


First-stage regression of all_road_2010_rays:

Statistics robust to heteroskedasticity and clustering on province05
Number of obs =                    257
Number of clusters (province05) =     26
---------------------------------------------------------------------------------
                |               Robust
all_road_2010~s |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
 road_1962_rays |   .3719791   .0834831     4.46   0.000     .2075496    .5364086
po_s_road_196.. |  -.6272756   .4261487    -1.47   0.142    -1.466624    .2120732
    ruralMigp00 |   2.14e-06   1.10e-06     1.95   0.052    -2.24e-08    4.30e-06
     larea_cc90 |     .10878   .1268168     0.86   0.392    -.1410002    .3585603
     larea_pf05 |   .0404845   .2041612     0.20   0.843    -.3616344    .4426034
province_capi~s |   1.440448   .5195824     2.77   0.006     .4170714    2.463826
lcensuspop198~f |     .76626   .1771551     4.33   0.000     .4173328    1.115187
frac_hig~2_pref |   6.100245   2.903858     2.10   0.037      .380763    11.81973
sha~an1982_pref |  -4.199371   2.060668    -2.04   0.043    -8.258093    -.140649
          _cons |  -9.959048   2.599649    -3.83   0.000    -15.07936    -4.83874
---------------------------------------------------------------------------------
F test of excluded instruments:
  F(  3,    25) =     8.90
  Prob > F      =   0.0003
Sanderson-Windmeijer multivariate F test of excluded instruments:
  F(  1,    25) =    28.86
  Prob > F      =   0.0000



Summary results for first-stage regressions
-------------------------------------------

                                           (Underid)            (Weak id)
Variable     | F(  3,    25)  P-val | SW Chi-sq(  1) P-val | SW F(  1,    25)
all_road_201 |       8.90    0.0003 |       31.11   0.0000 |       28.86
po_s_all_roa |       8.45    0.0005 |       11.79   0.0006 |       10.94
D_censuspop9 |       2.80    0.0604 |       32.41   0.0000 |       30.07

NB: first-stage test statistics cluster-robust

Stock-Yogo weak ID F test critical values for single endogenous regressor:
                                    5% maximal IV relative bias    13.91
                                   10% maximal IV relative bias     9.08
                                   20% maximal IV relative bias     6.46
                                   30% maximal IV relative bias     5.39
                                   10% maximal IV size             22.30
                                   15% maximal IV size             12.83
                                   20% maximal IV size              9.54
                                   25% maximal IV size              7.80
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for i.i.d. errors only.

Underidentification test
Ho: matrix of reduced form coefficients has rank=K1-1 (underidentified)
Ha: matrix has rank=K1 (identified)
Kleibergen-Paap rk LM statistic          Chi-sq(1)=7.26     P-val=0.0070

Weak identification test
Ho: equation is weakly identified
Cragg-Donald Wald F statistic                                       3.90
Kleibergen-Paap Wald rk F statistic                                 4.05

Stock-Yogo weak ID test critical values for K1=3 and L1=3:
                                                               <not available>

Weak-instrument-robust inference
Tests of joint significance of endogenous regressors B1 in main equation
Ho: B1=0 and orthogonality conditions are valid
Anderson-Rubin Wald test           F(3,25)=        3.81     P-val=0.0223
Anderson-Rubin Wald test           Chi-sq(3)=     12.33     P-val=0.0063
Stock-Wright LM S statistic        Chi-sq(3)=     11.57     P-val=0.0090

NB: Underidentification, weak identification and weak-identification-robust
    test statistics cluster-robust

Number of clusters             N_clust  =         26
Number of observations               N  =        257
Number of regressors                 K  =         10
Number of endogenous regressors      K1 =          3
Number of instruments                L  =         10
Number of excluded instruments       L1 =          3

IV (2SLS) estimation
--------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on province05

Number of clusters (province05) =     26              Number of obs =      257
                                                      F(  9,    25) =    15.45
                                                      Prob > F      =   0.0000
Total (centered) SS     =  24.19405425                Centered R2   =   0.2871
Total (uncentered) SS   =   67.2418726                Uncentered R2 =   0.7435
Residual SS             =  17.24765903                Root MSE      =    .2591

---------------------------------------------------------------------------------
                |               Robust
D_censu~9010_cp |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
all_road_2010~s |  -.0586738   .0259108    -2.26   0.024    -.1094581   -.0078895
po_s_all_road.. |  -.2520005   .1110736    -2.27   0.023    -.4697009   -.0343002
D_cen~9010_pref |   .7751506   .1733233     4.47   0.000     .4354431    1.114858
     larea_cc90 |  -.1661514   .0335645    -4.95   0.000    -.2319367   -.1003661
     larea_pf05 |   .0566459    .024906     2.27   0.023      .007831    .1054608
province_capi~s |   .2232924   .0737552     3.03   0.002     .0787349    .3678499
lcensuspop198~f |   .1387351   .0443026     3.13   0.002     .0519037    .2255666
frac_hig~2_pref |  -.2464789   .4256799    -0.58   0.563    -1.080796    .5878383
sha~an1982_pref |   -.288414   .2904417    -0.99   0.321    -.8576692    .2808413
          _cons |  -.7697138   .5826105    -1.32   0.186    -1.911609    .3721819
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):              7.264
                                                   Chi-sq(1) P-val =    0.0070
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):                3.896
                         (Kleibergen-Paap rk Wald F statistic):          4.054
Stock-Yogo weak ID test critical values:                       <not available>
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         all_road_2010_rays po_s_all_road_2010_ringxda
                      D_censuspop9010_pref
Included instruments: larea_cc90 larea_pf05 province_capitalPlus
                      lcensuspop1982_pref frac_highedu1982_pref
                      share_emp_man1982_pref
Excluded instruments: road_1962_rays po_s_road_1962_ringxda ruralMigp00
------------------------------------------------------------------------------

. outreg2 using "table4", excel  dec(4)  auto(2) symbol(***,**,*)  se addstat(Firs
> t stage F, e(widstat)) append
table4.xml
dir : seeout

. 
. 
. 
. log close
      name:  <unnamed>
       log:  C:\research\china\decentralization\restat_data\results\table4.log
  log type:  text
 closed on:  22 Jul 2016, 09:37:48
----------------------------------------------------------------------------------
